Pim van Geffen - Orebody Intelligence: Amalgamating Data Science and Orebody Knowledge

What if we could predict metallurgical and environmental performance using exploration data? And what if we could incorporate energy requirements, recovery performance, and closure costs into block models at the pre-feasibility stage? The quality of analytical data has greatly improved over the past few decades. At the same time, the power of data analytics has truly surged in recent years due to accessible data tools and leaps in computational capacity. These advancements allow us to extract significantly more value from our data. Now that data quality and analytics are no longer obstacles, it's only a matter of time before organizations shift their focus towards aligning all interconnected processes along the mining value chain, ultimately facilitating full-system optimization. As a small step in this direction, Pim van Geffen from ERM has developed workflows for material characterization to map, predict, and simulate downstream processes. If we start by selecting the right materials for metallurgical and environmental testwork, we can achieve greatest confidence in its results. And by establishing robust relationships between testwork results and drill-hole data, we can translate performance proxies into resource block models.